Men who have sex with men are the risk population most heavily impacted by HIV in the United States, and there substantial growth in the epidemic among young MSM. Given patterns of expanding access to the internet by Americans of color, and the high coverage among young Americans of social networking communities, the internet is a promising venue both for conducting sexual health research and delivering HIV prevention interventions to young MSM. However, despite some substantial advantages of the internet, including such as speed of data accrual and cost efficiency, there are also important limitations to internet-based data collections. For example, there is a lack of empirical data about the effectiveness of informed consent processes online, there is often substantial dropout from online surveys, and there are concerns that some people might fraudulently enroll in studies when monetary incentives are offered, compromising the validity of research results. Thus, we propose to conduct a series of experiments that will result in improved knowledge of how to conduct internet-based HIV prevention research with young MSM in ways that improve the protection of human subjects and decrease biases in data collections. The design will consist of two cross-sectional randomized controlled trials of a total of 2650 MSM recruited through social networking sites and gay-oriented Internet venues. In the first trial, we will test two alternative approaches to informed consent (video consent and FAQ-style bulleted consent) versus provision of a standard consent document. Outcomes will be comprehension of the content of the consent, and proportion of men who consent. In the second trial, we will develop and test non-monetary incentives to promote completion of surveys. Participants will be randomized to one of five arms: three kinds of non-monetary incentives (altruistic messaging, access to exclusive topical video content, and access to a dashboard, comparing the participant's responses to those of other MSM), a modest monetary incentive, or no incentive (standard of practice arm). The outcome will be completion of the survey, operationalized as time to failure in the survey. We expect that the proposed research will help support the identification and promotion of best methods for informed consent and engagement in research in this heavily impacted group of young Americans.